1. Field of the Invention
The present invention relates to LED print heads for electrophotographic printers and, more particularly, to a print head in which the use of light generated from an LED array is maximized. Specifically, the invention relates to an improvement in the light generated from an LED array with a so-called xe2x80x9cGRINxe2x80x9d (Gradient INdex) imaging lens as conventionally used in electrophotographic printers.
2. Description of the Related Art
Electrophotography is a well-known method of digital printing. In this process, a metal drum coated with a photoconductive material is employed. The photoconductor has the property that it is an insulator when no light falls on it, but when exposed to light, the material becomes conducting. In use, a surface charge is produced on the photoconductive material by, for example, placing a wire held at a high potential near the surface of the drum. By the action of light falling on the drum, parts of the surface of the drum can be discharged. The mechanism for the discharge is that, as a result of the light falling on the photoconductive material, the illuminated regions, which become conducting, allow the charge to leak away to the layer below the photoconductor which is typically held at earth potential. As a result of this process, a so-called latent image is created on the drum in the unexposed areas. Charged toner particles brought into the vicinity of the latent image are attracted to the latent image by electrostatic action. This results in a real image being created. This real image can then be transferred to a substrate, such as paper, by bringing the substrate into contact with the toned image.
The process described above is well-known in the literature and more detailed descriptions can be found in Schaffert, Electrophotography, Enlarged and Revised Edition, London, 1975, for example.
The latent image referred to above can be generated by a variety of optical means, such as a scanned laser, a liquid crystal light valve with a conventional light source, or an array of light-emitting diodes. Methods employing arrays of light emitting diodes (LEDs) are particularly attractive for high-speed applications where the data rate is a significant consideration, for wide imaging arrays where it would be impractical to use scanned lasers or where a particularly compact arrangement is required. In addition, since LED arrays are solid state, they offer some mechanical advantages over scanned lasers.
One of the constraints when considering conventional LEDs as a means of creating a latent image on a photoconductor for electrophotographic imaging is that the light emitted by the LED is essentially Lambertian (i.e., it is emitted into 2 xcfx80 steradians). Using the conventional imaging means (a GRIN lens array), it is only possible to collect a small portion of the emitted light and image it onto the photoconductor. A GRIN lens array consists of a row or rows of gradient index lenses. One particular type of array is a SELFOC(copyright) array which is a unique type of imaging device supplied by NSG in Japan and in which the rays of light passing through the lens follow sinusoidal paths from the front surface to the back surface of the lens. Typically, two rows of graded index lenses will be used with the two rows offset from each other by half the diameter of the lens (as in a hexagonal close packed structure). In contrast to a conventional lens, which relies on the curvature of its faces to refract and hence focus light, a gradient index lens is a rod of glass in which the refractive index varies according to the radial distance from the longitudinal axis of the rod. In a conventional lens, the image created by the lens is inverted. This means that in order to image an array of light-emitting diodes onto a photoconductor, it is necessary to have one imaging lens aligned with the axis of each light-emitting diode. However, a gradient index lens does not invert the image with respect to the object. It is, therefore, not a requirement to have a one-to-one match between each of the light-emitting diodes and the lens, since the image produced by one lens lines up exactly with that of another lens.
As with any lens, there needs to be a defined separation between the source and the imaging lens. For a conventional LED, with a Lambertian emission profile, this separation means that, typically, the amount collected by a SELFOC(copyright) array is around 3%. As a result of this low efficiency, most of the light generated is wasted. For high-speed applications, in particular, this inefficiency also means that unwanted heat is generated in the LEDs and the drive circuitry.
The present invention solves, or at least reduces, some or all of the aforementioned problems.
According to the present invention, there is provided a light-emitting diode (LED) print head, comprising a linear array of LEDs, a means of delivering current to each LED individually, and an imaging lens system to create a real image of the LED array, the image lens system including a plurality of gradient index lenses, characterized in that the LEDs each have emission cones closely matched to the numerical aperture of the gradient index imaging lenses. Preferably, this is achieved by the use of resonant cavity light-emitting diodes (RCLEDs) as the LEDs. The numerical aperture NA is conventionally defined as 1/f where f is the f-number of the lens as quoted in the manufacturer""s datasheets. The means of delivering the current to each individual light-emitting diode is preferably programmable through software to allow reconfiguration of the method of driving the print head. The data is preferably delivered to the print head via an optical fiber interface which has the advantages of noise immunity and bandwidth increase. Data is provided to drive electronics used to control the intensity or pulse length of printing pulses and to initiate firing of the LEDs in the array at the appropriate times. Preferably, calibration data is delivered to the print head via an optical fiber interface. This avoids the need to store the calibration data locally at the print head. The data to control the brightness of the light-emitting diodes are preferably delivered via an optical fiber interface to compensate for changes in the print speed.
Although there are a variety of methods of controlling the emission angle of LEDs to be less than 2 xcfx80 steradians, including attaching a microlens to each light-emitting diode, many of these are not particularly efficient. However, one advantageous method of controlling the emission angle is to use resonant cavity light-emitting diode (RCLED) technology. In this technology, an active light-emitting area is sandwiched between two mirrors that form a Fabry Perot resonator. The Fabry Perot resonator is designed such that the emission from the active layer lies within the pass band of the resonator. In detail, a number of layers of semiconductor material are deposited on a gallium arsenide substrate by metal-organic vapor phase epitaxy (MOVPE) to form a first Bragg reflector. Typically, these incorporate N-type doping. Next, the light-emitting layer is deposited typically in the form of gallium indium phosphide quantum wells with aluminum gallium indium phosphide barriers. This active quantum well layer is designed to have an optical thickness of one wavelength. A second Bragg structure is then deposited onto the quantum well layer, but this time with a P-type of doping, so that a P-N junction is formed between the two mirrors. A highly doped layer of gallium arsenide will typically be deposited onto the top of the structure to allow electrical contact to be formed.
The degree to which the emission profile is modified from being Lambertian, which would be the case without the Bragg mirrors, towards a pencil beam is a function of the reflectivity of the Bragg mirrors and the degree of detuning between the Fabry Perot resonance and the center of the quantum well emission. By selecting the correct reflectivity for the Bragg reflectors and the appropriate level of detuning, the emission profile for the LED can be matched to the collection aperture of the GRIN imaging lens.
The modification of the emission profile is controlled by the MOVPE deposition process on the wafer. Individual LEDs can be created on the wafer by using photolithography and etching to isolate individual light-emitting diodes from their neighbors. The deposition of insulating layers and metallization are used to produce the electrical contacts to the LED.
By utilizing this new LED technology, it is possible to achieve an improvement of more than 20 times in the useful light generated from an LED when used with a SELFOC(copyright) imaging lens. This improvement is illustrated in the graphs in FIGS. 6 and 7. By way of example, if we consider SELFOC(copyright) array type SLA 12 produced by NSG, then the distance from the light source to the input side of the lens array is 7.89 mm. Each lens in the array has a restricted view angle. In the case of SLA 12, this is 23.34 degrees (full angle). Therefore, any light that is emitted in a cone angle greater than this will not be collected by the SELFOC(copyright) at all. In actual fact, the situation is slightly worse than this in that since there are only two rows of lenses not all of the light that is emitted into this cone angle will be collected. In order to ensure that all of the emitted light is collected, the emitted cone angle should be less than 15.8 degrees full angle.